Optimising the timing of food intake relative to exercise may maximise the effectiveness of free-living exercise programmes on improvements in glycaemic control and cardio-metabolic health. This study aimed to assess the feasibility of a free-living, walking-based exercise programme and determine whether undertaking each exercise session before or after breakfast would most benefit longer-term metabolic health. Thirty-four people living with obesity (43±12 y, BMI 35.1±5.1 kg.m-2) undertook a 12-week walking-based programme, consisting of two continuous (30-60 min at 50% HRmax) and two interval exercise sessions per week (30-60 min, alternating 3 min at 85% HRmax and 3 min at 50% HRmax). Participants were allocated to exercise before (FASTED) or after (FED) breakfast (n = 17 per group). Feasibility (acceptability, adherence and compliance) to the exercise intervention were assessed, as well as changes in anthropometric variables, 24-hour continuous glucose monitoring, serum biochemistry including HbA1c, lipid profile and liver transaminases. Exercise adherence (FASTED: 93±4%, FED: 95±5%) and compliance (FASTED: 85±10%, FED: 88±10%) was high in both groups, and participants described exercise monitoring, programme structure and support as facilitators to this. Body mass, BMI, waist-to-hip ratio and HbA1c decreased similarly between groups (all P<0.01). However, serum ALT concentrations decreased after FASTED (-16± -14%; P = 0.001), but not FED training (-2 ± -4%; P = 0.720). We demonstrate that a free-living walking-based exercise programme, with exercise timed relative to breakfast can achieve high adherence and compliance and improve some anthropometric variables and HbA1c. Whether FASTED exercise can elicit greater improvements in liver health requires further investigation.